Alexandre, thanks for sharing your insights. How would the numbers change, if you would assign values of -1, -0.5, 0.5 and 1 to the four categories? Treating "some" and "a lot" the same (1) seems like a strange decision to me. Furthermore, the dataset link only leads to a PEW article that does not seem to yield the partisan data that you have processed. Can you please guide me more specifically to the used data? Is there maybe even a git repo that you would want to share?

Expand full comment